Tumor development is dependent upon the inactivation of two key tumor suppressor pathways, p16- cycD/cdk4-pRB-E2F and p19Arf-mdm2-p53, that together regulate cellular proliferation and the tumor surveillance response. This proposal focuses on one member of the E2f family, E2f3, which is a component of both the pRB and p53 networks and whose amplification has been linked to the development of human bladder and prostate tumors. E2f3 encodes two distinct proteins, called E2F3A and E2F3B that differ only in their N terminal sequences. Existing studies do not address whether the observed up-regulation of E2F3 in human tumors reflects an increase in the levels of E2F3A, E2F3B or both isoforms. Thus, it is currently unclear whether either, or both, of these proteins contribute to tumor development. The analysis of mutant mouse models has provided unequivocal proof that E2f3 acts in dose-dependent manner to promote tumor formation in vivo. This validates the mouse as model system to investigate E2f3's oncogenic activity. Preliminary studies suggest that the E2F3A and E2F3B proteins have distinct biological properties in vivo. E2F3A is believed to promote cellular proliferation by activating the transcription of genes that encode key components of the cell cycle control machinery. In contrast, E2F3B contributes to the transcriptional repression of the Art tumor suppressor and thereby impedes induction of the p53 tumor surveillance response. The goal of this proposal is to test the hypothesis that E2F3A and E2F3B have differential roles in vivo and to determine how each of these isoforms contribute to E2f3's roles in cellular proliferation, normal development and tumorigenicity. This proposal has three aims: (1) To identify the relative roles of E2F3A and E2F3B in cellular processes; (2) To use mutant mouse strains to determine the role of E2F3A and E2F3B in tumorigenesis; and (3) To investigate the mechanism of action of the E2F3 repressor by elucidating the components of the E2F3 repressor complex and identifying its specific downstream target genes. [unreadable] [unreadable] [unreadable]